The proposed study will be based on the continuation and expansion of the culturally-appropriate smoking cessation community intervention begun with San Francisco Hispanics in 1988. The community intervention has been based on promoting smoking cessation among current smokers and motivating non-smokers to help smokers quit. Free distribution of a Spanish language self-help smoking cessation guide (Guia Para Dejar de Fumar) has been the principal intervention component. The intervention has made wide use of electronic and printed media, established a presence in the community, offered free smoking cessation groups, and promoted available services to health providers practicing in the Hispanic community. We now propose continuation of the intervention for five more years with an expansion of its components to include prevention of smoking initiation among Hispanic adolescents and prevention of smoking relapse among smokers who quit. In order to develop culturally appropriate smoking prevention interventions targeting Hispanic adolescents, we will conduct a cross-cultural study to identify factors involved in initiation of smoking. This study will compare responses of 300 Hispanic and 150 White non-hispanic adolescents to a structured questionnaire addressing smoking experimentation and initiation. Data being collected now from smokers who quit and subsequently have relapsed will be analyzed in order to develop a more intense relapse prevention component. The association of increased depressive symptoms and current smoking, especially among Hispanic women, will also need to be addressed in the expansion of the intervention. The community intervention will be evaluated by annual cross-sectional surveys of 2000 Hispanics in San Francisco with comparisons to two baseline surveys completed in 1986-- 87. Given that the content of intervention is primarily in Spanish, less acculturated Hispanics will also be compared to more acculturated Hispanics. A cohort of current and former smokers will be recruited from the cross-sectional samples and followed prospectively with annual interviews to determine smoking behavior. Self-reported quitters will be validated with saliva cotinine.